Strava activities - Analysis, investigation and visualization of two years of recorded efforts

# to avoid geopandas warnings, don't run if you don't mind warnings
import warnings
warnings.filterwarnings('ignore')

import geopandas as gpd
import pandas as pd
import numpy as np
import pyrosm
import matplotlib.pyplot as plt
import gpxpy
from geopy.geocoders import Nominatim
import os
from tqdm import tqdm
import movingpandas as mpd
import utils
from datetime import timedelta
import contextily as ctx
from sklearn.cluster import dbscan
import json
import folium
from folium import plugins

Data loading

activities = []
runs = []
hikes = []
rides = []

for file in tqdm(os.listdir("data/strava_activities/")):
    gpx_file = open("data/strava_activities/{}".format(file), 'r')
    gpx = gpxpy.parse(gpx_file)
    if((gpx.tracks[0].type == 'Run') | (gpx.tracks[0].type == 'running') ):
        runs.append(gpx)
        activities.append(gpx)
    elif(gpx.tracks[0].type == 'Hike'):
        hikes.append(gpx)
        activities.append(gpx)
    elif(gpx.tracks[0].type == 'Ride'):
        rides.append(gpx)
        activities.append(gpx)

100%|██████████| 159/159 [00:12<00:00, 12.93it/s]

print("Total number of activities: \t{}".format(len(activities)))
print("Total number of runs: \t\t{}".format(len(runs)))
print("Total number of rides: \t\t{}".format(len(rides)))
print("Total number of hikes: \t\t{}".format(len(hikes)))

Total number of activities: 	152
Total number of runs: 		135
Total number of rides: 		3
Total number of hikes: 		14

Which are the longest activities in term of time and length? Let's pick the top 5 for both aspects.

utils.getTopLongestTravelTime(activities, 5)
print("-----------------------------")
utils.getTopLongestTravel(activities, 5)
print("-----------------------------")
utils.getTopElevationDifference(activities, 5)

1° longest activity duration: 10:26:45, activity n° 91, type: Ride
2° longest activity duration: 09:37:45, activity n° 125, type: Hike
3° longest activity duration: 08:33:00, activity n° 15, type: Hike
4° longest activity duration: 07:59:45, activity n° 121, type: Hike
5° longest activity duration: 05:19:14, activity n° 7, type: Hike
-----------------------------
1° longest activity length: 102401.12, activity n° 91, type: Ride
2° longest activity length: 35057.82, activity n° 5, type: Ride
3° longest activity length: 25111.91, activity n° 45, type: Ride
4° longest activity length: 21448.31, activity n° 109, type: Run
5° longest activity length: 19784.85, activity n° 15, type: Hike
-----------------------------
1° highest elevation difference: 1489.0, activity n° 15, type: Hike
2° highest elevation difference: 1368.6, activity n° 7, type: Hike
3° highest elevation difference: 1266.0, activity n° 121, type: Hike
4° highest elevation difference: 969.6, activity n° 115, type: Hike
5° highest elevation difference: 961.6, activity n° 84, type: Hike

# list of dataframes of all the activities
activities_dfList = utils.toList(activities)
runs_dfList = utils.toList(runs)
rides_dfList = utils.toList(rides)
hikes_dfList = utils.toList(hikes)

100%|██████████| 152/152 [00:04<00:00, 33.27it/s]
100%|██████████| 135/135 [00:02<00:00, 60.68it/s]
100%|██████████| 3/3 [00:00<00:00, 20.10it/s]
100%|██████████| 14/14 [00:00<00:00, 22.48it/s]

Let's check start and arrival points of my longest trip, the activity n° 91.

start = activities_dfList[91][:1]
end = activities_dfList[91][-1:]

latlon = str(start.latitude[0]) + "," + str(start.longitude[0])
geolocator = Nominatim(user_agent="geospatial course unitn")
starting_place = geolocator.reverse(latlon)
latlon = str(end.latitude.values[0]) + "," + str(end.longitude.values[0])
geolocator = Nominatim(user_agent="geospatial course unitn")
arrival_place = geolocator.reverse(latlon)
print("Starting place: {}".format(starting_place))
print("Arrival place: {}".format(arrival_place))

Starting place: 14, Via Vittorio Veneto, Clarina, San Pio X, Trento, Territorio Val d'Adige, Provincia di Trento, Trentino-Alto Adige/Südtirol, 38122, Italia
Arrival place: Via Don Narciso Sordo, Clarina, San Pio X, Trento, Territorio Val d'Adige, Provincia di Trento, Trentino-Alto Adige/Südtirol, 38122, Italia

Starting and arrival point are quite near. This could be a ring trip, but let's check it out by plotting it.

# Set time as index for movingpandas
for i in range(len(activities_dfList)):
    activities_dfList[i].set_index('time', drop=True, inplace=True)
for i in range(len(runs_dfList)):
    runs_dfList[i].set_index('time', drop=True, inplace=True)
for i in range(len(rides_dfList)):
    rides_dfList[i].set_index('time', drop=True, inplace=True)
for i in range(len(hikes_dfList)):
    hikes_dfList[i].set_index('time', drop=True, inplace=True)

# List of geodataframes of activities
geo_dfList = utils.toGdfList(activities_dfList)
runs_geo_dfList = utils.toGdfList(runs_dfList)
rides_geo_dfList = utils.toGdfList(rides_dfList)
hikes_geo_dfList = utils.toGdfList(hikes_dfList)

100%|██████████| 152/152 [00:03<00:00, 38.72it/s]
100%|██████████| 135/135 [00:03<00:00, 38.00it/s]
100%|██████████| 3/3 [00:00<00:00, 43.30it/s]
100%|██████████| 14/14 [00:00<00:00, 40.29it/s]

# List of all trajectories of the dataset
trajectories = utils.getTrajList(geo_dfList)
runsTrajectories = utils.getTrajList(runs_geo_dfList)
ridesTrajectories = utils.getTrajList(rides_geo_dfList)
hikesTrajectories = utils.getTrajList(hikes_geo_dfList)

Activity visualization(s)

trajectories[91].plot()
plt.show()

trajectories[91].hvplot(geo=True, tiles='EsriImagery', line_width=5, color='lightblue')

So apparently this wasn't a ring trip, but an a/r trip to Garda Lake. Let's investigate more about it.

for i in tqdm(range(len(trajectories))):
    trajectories[i].add_speed(overwrite=True)
for i in tqdm(range(len(trajectories))):
    trajectories[i].df['kmh'] = trajectories[i].df['speed'].apply(utils.ms_to_km)

100%|██████████| 152/152 [00:44<00:00,  3.44it/s]
100%|██████████| 152/152 [00:00<00:00, 604.90it/s]

trajectories[91].df

	longitude	latitude	elevation	geometry	speed	kmh
time
2021-04-24 07:43:50	11.119156	46.057613	263.4	POINT Z (11.11916 46.05761 263.40000)	1.618042	5.824950
2021-04-24 07:43:53	11.119101	46.057592	263.2	POINT Z (11.11910 46.05759 263.20000)	1.618042	5.824950
2021-04-24 07:44:01	11.118860	46.057567	261.4	POINT Z (11.11886 46.05757 261.40000)	2.356896	8.484827
2021-04-24 07:44:08	11.118716	46.057441	248.0	POINT Z (11.11872 46.05744 248.00000)	2.556767	9.204360
2021-04-24 07:44:10	11.118725	46.057395	242.8	POINT Z (11.11872 46.05739 242.80000)	2.580113	9.288407
...	...	...	...	...	...	...
2021-04-24 18:10:26	11.119483	46.056766	182.4	POINT Z (11.11948 46.05677 182.40000)	4.510490	16.237765
2021-04-24 18:10:27	11.119526	46.056744	182.2	POINT Z (11.11953 46.05674 182.20000)	4.129420	14.865911
2021-04-24 18:10:30	11.119630	46.056667	182.0	POINT Z (11.11963 46.05667 182.00000)	3.916085	14.097904
2021-04-24 18:10:33	11.119744	46.056621	182.0	POINT Z (11.11974 46.05662 182.00000)	3.398808	12.235708
2021-04-24 18:10:35	11.119790	46.056612	182.0	POINT Z (11.11979 46.05661 182.00000)	1.848786	6.655629

3122 rows × 6 columns

print("Average pace: {}; max pace: {}".format(np.round(np.average(trajectories[91].df.kmh), 2), np.round(np.max(trajectories[91].df.kmh), 2)))

Average pace: 20.16; max pace: 51.01

20 km/h as average pace isn't so bad, but 51 km/h it's definitely out of my capabilities. Where can I have had that peak?

trajectories[91].hvplot(c='kmh',  geo=True, tiles='OSM', cmap='Reds', line_width=5, colorbar=True)

We can see that from Nago to Torbole there is a street portion in which I had a huge pace in a direction, but a really low one in the other direction. Why that?

highSpeed = trajectories[91].df[trajectories[91].df.kmh > 40.0]
highSpeed

	longitude	latitude	elevation	geometry	speed	kmh
time
2021-04-24 10:44:55	10.885169	45.877044	188.0	POINT Z (10.88517 45.87704 188.00000)	11.440208	41.184750
2021-04-24 10:45:00	10.884649	45.876551	184.0	POINT Z (10.88465 45.87655 184.00000)	13.612350	49.004460
2021-04-24 10:45:05	10.884096	45.876046	177.4	POINT Z (10.88410 45.87605 177.40000)	14.133370	50.880133
2021-04-24 10:45:10	10.883541	45.875546	169.6	POINT Z (10.88354 45.87555 169.60000)	14.064301	50.631484
2021-04-24 10:45:15	10.883008	45.875038	162.8	POINT Z (10.88301 45.87504 162.80000)	14.000708	50.402549
2021-04-24 10:45:19	10.882568	45.874675	157.2	POINT Z (10.88257 45.87468 157.20000)	13.216540	47.579545
2021-04-24 10:45:23	10.882104	45.874306	151.6	POINT Z (10.88210 45.87431 151.60000)	13.647074	49.129467
2021-04-24 10:45:25	10.881868	45.874115	148.8	POINT Z (10.88187 45.87412 148.80000)	14.021509	50.477434
2021-04-24 10:45:27	10.881623	45.873926	145.8	POINT Z (10.88162 45.87393 145.80000)	14.169672	51.010819
2021-04-24 10:45:30	10.881269	45.873657	141.6	POINT Z (10.88127 45.87366 141.60000)	13.537351	48.734462
2021-04-24 10:45:31	10.881156	45.873570	140.2	POINT Z (10.88116 45.87357 140.20000)	13.056724	47.004207
2021-04-24 10:45:33	10.880917	45.873396	137.2	POINT Z (10.88092 45.87340 137.20000)	13.401048	48.243773
2021-04-24 10:45:35	10.880679	45.873223	134.8	POINT Z (10.88068 45.87322 134.80000)	13.334094	48.002739
2021-04-24 10:45:37	10.880451	45.873061	132.6	POINT Z (10.88045 45.87306 132.60000)	12.625165	45.450593
2021-04-24 10:45:43	10.879825	45.872616	126.0	POINT Z (10.87983 45.87262 126.00000)	11.557422	41.606719
2021-04-24 10:45:46	10.879511	45.872388	122.8	POINT Z (10.87951 45.87239 122.80000)	11.721599	42.197757
2021-04-24 10:45:49	10.879201	45.872153	119.6	POINT Z (10.87920 45.87215 119.60000)	11.839487	42.622153
2021-04-24 10:45:52	10.878904	45.871904	117.4	POINT Z (10.87890 45.87190 117.40000)	12.007906	43.228463
2021-04-24 10:45:58	10.878309	45.871429	110.8	POINT Z (10.87831 45.87143 110.80000)	11.692301	42.092285
2021-04-24 10:46:04	10.877811	45.870949	104.0	POINT Z (10.87781 45.87095 104.00000)	11.440576	41.186072
2021-04-24 10:46:06	10.877681	45.870762	101.6	POINT Z (10.87768 45.87076 101.60000)	11.553034	41.590921
2021-04-24 15:03:47	10.908134	45.871694	243.8	POINT Z (10.90813 45.87169 243.80000)	11.178470	40.242492

Excluding the last point, all the others are sequential. This indicates a particular portion of the route where I reached that pace. Could you guess the reason?

print("Segment length: {} m; segment elevation change: {} m".format(np.round(highSpeed[0:1].to_crs('epsg:32632').geometry[0].distance(highSpeed[-2:].to_crs('epsg:32632').geometry[0]), 3), highSpeed[-2:].elevation[0] - highSpeed[0:1].elevation[0]))

Segment length: 908.454 m; segment elevation change: -86.4 m

That's why: 86m of height difference in less than a km, it's a quite steep descent! But truth be told, I still have clear memories about the ascent on our way back..

Activities visualization - first attempt

Plot all hike activities in a map. HVplot guarantees a pleasant visualization, but with hikes it's pointless to find clusters of activities. In any case, this is a good example of how trajectories can be mapped together and be easily identified when not overlapped.

hikesToPlot = mpd.TrajectoryCollection(hikesTrajectories)

hikesToPlot.hvplot(line_width=3, geo=True, tiles='EsriImagery')

# about 7-8 mins to render
all = mpd.TrajectoryCollection(trajectories)
#all.hvplot(line_width=3, geo=True, tiles='EsriImagery')

On the other hand, with too many overlapping activities it's complicated to identify how many of them are in a given area (even colours could be misleading). We can isolate starting points and plot them to have a better idea.

starting_points = all.get_start_locations().set_crs('epsg:4326')
starting_points.drop(columns={'kmh', 'speed'}, inplace=True)
starting_points

	elevation	geometry	latitude	longitude
2019-06-06 18:13:25	198.6	POINT Z (11.12007 46.05674 198.60000)	46.056745	11.120075
2020-11-19 19:06:51	198.2	POINT Z (11.11502 46.05746 198.20000)	46.057457	11.115020
2021-05-08 10:08:24	78.4	POINT Z (10.87843 45.92002 78.40000)	45.920023	10.878429
2021-05-04 17:24:42	195.4	POINT Z (11.11215 46.05954 195.40000)	46.059543	11.112147
2020-05-17 19:17:09	203.8	POINT Z (11.11181 46.05405 203.80000)	46.054046	11.111814
...	...	...	...	...
2019-07-30 18:24:13	188.0	POINT Z (11.11154 46.05242 188.00000)	46.052419	11.111539
2021-01-10 18:50:04	189.2	POINT Z (11.11213 46.05954 189.20000)	46.059536	11.112131
2020-05-03 19:12:34	201.2	POINT Z (11.11157 46.05244 201.20000)	46.052438	11.111570
2021-03-25 19:42:42	216.6	POINT Z (11.12668 46.05870 216.60000)	46.058704	11.126684
2021-04-21 18:04:03	202.0	POINT Z (11.11940 46.05786 202.00000)	46.057857	11.119398

152 rows × 4 columns

With a scatter plot of the starting points we can have an idea of how they are distributed.

ax = starting_points.plot(figsize=(10,10), color='xkcd:cobalt blue')
ctx.add_basemap(ax, crs=starting_points.crs.to_string(), source=ctx.providers.OpenStreetMap.Mapnik)

Clustering activities

starting_points['x'] = starting_points.to_crs(epsg=3857).geometry.x
starting_points['y'] = starting_points.to_crs(epsg=3857).geometry.y

bounding_box = [starting_points.x.min(), starting_points.y.min(), starting_points.x.max(), starting_points.y.max()]
basemap, basemap_extent = ctx.bounds2img(*bounding_box, zoom=10,source = ctx.providers.Stamen.Terrain)
data_extent = [starting_points.x.min(), starting_points.x.max(), starting_points.y.min(), starting_points.y.max()]

utils.plotHex(starting_points, data_extent, basemap, basemap_extent, hex=25)

As predictable, most activities have been started from Trento. It would be pointless to carry out more analysis at this level, but we can check the distribution over the city area.

trentoBoundary = utils.get_boundary("Trento")

Getting boundaries for Trento
Extracted boundaries for Trento. Time elapsed: 35.04 s

starting_points_tn = gpd.GeoDataFrame()
for i in range(len(starting_points)):
    if trentoBoundary.geometry[0].contains(starting_points[:].geometry[i]):
        starting_points_tn = starting_points_tn.append(starting_points[i:i+1])
print("Attività in zona Trento: {}".format(len(starting_points_tn)))

Attività in zona Trento: 134

bounding_box = [starting_points_tn.x.min(), starting_points_tn.y.min(), starting_points_tn.x.max(), starting_points_tn.y.max()]
basemap, basemap_extent = ctx.bounds2img(*bounding_box, zoom=14,source = ctx.providers.Stamen.Terrain)
data_extent = [starting_points_tn.x.min(), starting_points_tn.x.max(), starting_points_tn.y.min(), starting_points_tn.y.max()]

utils.plotHex(starting_points_tn, data_extent, basemap, basemap_extent, hex=25)

It's possible to see a first pattern: most of the activities started in the south-west part of the city, near the Adige river. Is it possible to better understand this?

utils.plotKDE(starting_points_tn, data_extent, basemap, basemap_extent, grad=20)

This visualization confirm the main area of starting places. Can we infer something more with a clustering? Maybe the area where I live?

minp = np.round(starting_points_tn.shape[0]*0.15)   # 15% representation
clusterindexes, labels = dbscan(starting_points_tn[['x', 'y']], eps=300, min_samples=minp) # points within 300m
labels=pd.Series(labels, index=starting_points_tn.index)

utils.plotClusters(starting_points_tn, labels, basemap, basemap_extent)

Using a dbscan technique it's clear that there are three main areas of starting points: two belongs to the Lungadige, while the other is located in San Pio X neighbourhood. So, assuming that I'm not living on the river, I probably live in the third area found (which is a right guess).

Strava-like heatmap visualization

For a further visualization, let's check out an heatmap of all the activities loaded.

coords = []
for i in tqdm(range(len(geo_dfList))):
    for idx, row in geo_dfList[i].iterrows():
        lat = row.latitude
        lon = row.longitude
        coords.append( (lat, lon) )

100%|██████████| 152/152 [00:16<00:00,  9.06it/s]

mapFolium = folium.Map(location=[46.0672,11.1212], tiles='CartoDB dark_matter', zoom_start=12)
plugins.HeatMap(coords, radius=2.5, blur=1.5, min_opacity=0.3, max_value=1, gradient=utils.HEATMAP_GRAD['dark']).add_to(mapFolium)
mapFolium

Make this Notebook Trusted to load map: File -> Trust Notebook

Thanks to Folium, there is a representation with a heatmap of all the activities. It's easy to notice a greater concentration of points in the areas highlighted before.

Vigolana Hike - Analysis with OSM

Hiking activities could be interesting to analyze. For example, given that I'm not a good biker (or a good trail runner...yet), activities with an high elevation difference between highest and lowest points should be hikes.

utils.getTopElevationDifference(activities,5)

1° highest elevation difference: 1489.0, activity n° 15, type: Hike
2° highest elevation difference: 1368.6, activity n° 7, type: Hike
3° highest elevation difference: 1266.0, activity n° 121, type: Hike
4° highest elevation difference: 969.6, activity n° 115, type: Hike
5° highest elevation difference: 961.6, activity n° 84, type: Hike

trajectories[115].hvplot(c='elevation',  geo=True, tiles='OSM', cmap='Blues', line_width=5, colorbar=True)

Stops analysis

For example, activity n°115 has been a hike to the Vigolana Shelter (Madonnina) made last summer. Did I took any stop on my way up? Was them for any specific reason?

vigolana = trajectories[115]
stopDetector = mpd.TrajectoryStopDetector(vigolana)
stop_durations = stopDetector.get_stop_time_ranges(min_duration=timedelta(seconds=180), max_diameter=30)
stop_points = stopDetector.get_stop_points(min_duration=timedelta(seconds=180), max_diameter=30)

for i in stop_durations:
    print(i)

Traj 115: 2021-07-21 16:12:34 - 2021-07-21 16:15:40 (duration: 0 days 00:03:06)
Traj 115: 2021-07-21 16:23:13 - 2021-07-21 16:27:03 (duration: 0 days 00:03:50)
Traj 115: 2021-07-21 16:33:47 - 2021-07-21 16:37:31 (duration: 0 days 00:03:44)
Traj 115: 2021-07-21 16:38:45 - 2021-07-21 16:44:36 (duration: 0 days 00:05:51)
Traj 115: 2021-07-21 16:51:21 - 2021-07-21 16:59:48 (duration: 0 days 00:08:27)
Traj 115: 2021-07-21 17:14:49 - 2021-07-21 17:20:11 (duration: 0 days 00:05:22)
Traj 115: 2021-07-21 17:24:14 - 2021-07-21 17:29:49 (duration: 0 days 00:05:35)
Traj 115: 2021-07-21 17:38:10 - 2021-07-21 17:42:09 (duration: 0 days 00:03:59)

# Only eight stops in a short range for more than 3 minutes. Where were them?
stop_points

	geometry	start_time	end_time	traj_id	duration_s
stop_id
115_2021-07-21 16:12:34	POINT Z (11.19811 45.97597 1261.80000)	2021-07-21 16:12:34	2021-07-21 16:15:40	115	186.0
115_2021-07-21 16:23:13	POINT Z (11.19738 45.97392 1379.00000)	2021-07-21 16:23:13	2021-07-21 16:27:03	115	230.0
115_2021-07-21 16:33:47	POINT Z (11.19543 45.97252 1459.40000)	2021-07-21 16:33:47	2021-07-21 16:37:31	115	224.0
115_2021-07-21 16:38:45	POINT Z (11.19555 45.97222 1469.80000)	2021-07-21 16:38:45	2021-07-21 16:44:36	115	351.0
115_2021-07-21 16:51:21	POINT Z (11.19424 45.97240 1539.20000)	2021-07-21 16:51:21	2021-07-21 16:59:48	115	507.0
115_2021-07-21 17:14:49	POINT Z (11.18993 45.97007 1724.00000)	2021-07-21 17:14:49	2021-07-21 17:20:11	115	322.0
115_2021-07-21 17:24:14	POINT Z (11.18868 45.96932 1788.80000)	2021-07-21 17:24:14	2021-07-21 17:29:49	115	335.0
115_2021-07-21 17:38:10	POINT Z (11.18717 45.96960 1868.60000)	2021-07-21 17:38:10	2021-07-21 17:42:09	115	239.0

Let's see where these stops have been taken along the hike.

vigoPlot = vigolana.hvplot(title='Vigolana Hike', line_width=3, tiles='StamenTerrain', cmap='Blues', c='elevation', colorbar=True)
stopPlot = vigoPlot * stop_points.hvplot(geo=True, size='duration_s', color='red')
stopPlot

A possible cause for the stops could be the high elevation gain made in a short time. Let's check if it's the case.

stop_points = utils.getStopElevationDiff(stop_points, vigolana).set_crs(epsg=4326)
stop_points

	geometry	start_time	end_time	traj_id	duration_s	elevation_diff	time_diff
stop_id
115_2021-07-21 16:12:34	POINT Z (11.19811 45.97597 1261.80000)	2021-07-21 16:12:34	2021-07-21 16:15:40	115	186.0	231.0	0 days 00:20:17
115_2021-07-21 16:23:13	POINT Z (11.19738 45.97392 1379.00000)	2021-07-21 16:23:13	2021-07-21 16:27:03	115	230.0	117.2	0 days 00:07:33
115_2021-07-21 16:33:47	POINT Z (11.19543 45.97252 1459.40000)	2021-07-21 16:33:47	2021-07-21 16:37:31	115	224.0	80.4	0 days 00:06:44
115_2021-07-21 16:38:45	POINT Z (11.19555 45.97222 1469.80000)	2021-07-21 16:38:45	2021-07-21 16:44:36	115	351.0	10.4	0 days 00:01:14
115_2021-07-21 16:51:21	POINT Z (11.19424 45.97240 1539.20000)	2021-07-21 16:51:21	2021-07-21 16:59:48	115	507.0	69.4	0 days 00:06:45
115_2021-07-21 17:14:49	POINT Z (11.18993 45.97007 1724.00000)	2021-07-21 17:14:49	2021-07-21 17:20:11	115	322.0	184.8	0 days 00:15:01
115_2021-07-21 17:24:14	POINT Z (11.18868 45.96932 1788.80000)	2021-07-21 17:24:14	2021-07-21 17:29:49	115	335.0	64.8	0 days 00:04:03
115_2021-07-21 17:38:10	POINT Z (11.18717 45.96960 1868.60000)	2021-07-21 17:38:10	2021-07-21 17:42:09	115	239.0	79.8	0 days 00:08:21

In some cases stops could explained by elevation gain in a short amount of time, but in other there should be different reasons. A good viewpoint to take pictures while resting, maybe? Let's check with the longest stop I took.

longestStop = stop_points[stop_points.duration_s==max(stop_points.duration_s)]

Including OSM data for the analysis

bbox = vigolana.get_mcp()

# All point of interests in the minimum convex polygon containing the hike
hikeOSM = pyrosm.OSM("./data/altopiano_della_vigolana.osm.pbf", bounding_box=bbox)
hikePois = hikeOSM.get_pois()
hikePois

	lat	tags	version	changeset	lon	timestamp	id	name	operator	ref	drinking_water	bicycle	information	tourism	geometry	osm_type	amenity	building
0	45.972450	{"direction":"N"}	3	0.0	11.194252	1594922072	910787221	Polsa	None	None	None	None	None	viewpoint	POINT (11.19425 45.97245)	node	NaN	NaN
1	45.971054	{"ele":"2030","fee":"yes","capacity":"6"}	6	0.0	11.185942	1572786755	914756310	Bivacco Vigolana	Società Alpinisti Tridentini	None	no	None	None	wilderness_hut	POINT (11.18594 45.97105)	node	NaN	NaN
2	45.976658	{"mtb":"yes","horse":"yes","hiking":"yes"}	1	0.0	11.198959	1340702685	1802521175	None	None	None	None	no	guidepost	information	POINT (11.19896 45.97666)	node	NaN	NaN
3	45.979012	{"mtb":"yes","horse":"yes","hiking":"yes"}	1	0.0	11.201583	1340702687	1802521237	None	None	None	None	no	guidepost	information	POINT (11.20158 45.97901)	node	NaN	NaN
4	45.980946	{"mtb":"yes","horse":"yes","hiking":"yes"}	1	0.0	11.203117	1340702689	1802521309	None	None	None	None	no	guidepost	information	POINT (11.20312 45.98095)	node	NaN	NaN
5	45.982376	{"mtb":"yes","horse":"yes","hiking":"yes"}	1	0.0	11.203637	1340702689	1802521315	None	None	None	None	no	guidepost	information	POINT (11.20364 45.98238)	node	NaN	NaN
6	45.969315	{"hiking":"yes","wikimedia_commons":"File:Segn...	4	0.0	11.185840	1615634668	4437914625	None	None	425-444	None	None	guidepost	information	POINT (11.18584 45.96931)	node	NaN	NaN
7	NaN	NaN	2	NaN	NaN	1532186863	152409892	Bivacco Madonnina	NaN	NaN	NaN	NaN	NaN	NaN	POLYGON ((11.18591 45.97107, 11.18596 45.97107...	way	shelter	yes

for i in range(len(hikePois)):
    if hikePois.to_crs('epsg:32632').geometry.values[i].buffer(10).contains(longestStop.to_crs('epsg:32632').geometry.values[0]):
        print("Name: {}; tourism: {}".format(hikePois.at[i, 'name'], hikePois.at[i, 'tourism']))

Name: Polsa; tourism: viewpoint

Apparently, I stopped for a short break, but also to take a nice picture and enjoy the view. Let's confirm it in the map seen before, but changing the layer with a more appropriate one.

vigoPlot = vigolana.hvplot(title='Vigolana Hike', line_width=3, tiles='OSM', cmap='Blues', c='elevation', colorbar=True)
stopPlot = vigoPlot * longestStop.hvplot(geo=True, size='duration_s', color='red')
stopPlot

As expected, with the OSM layer we can see that the longest stop has been taken in proximity (within 10m) of the viewpoint Polsa.
Fun fact: this viewpoint has been mapped last time in 2020 by the user Martin Larcher (https://www.openstreetmap.org/node/910787221); at the end of the hike we just analyzed, I spent the night in the Madonnina shelter with...Martin Larcher himself, along with his family!

Performance analysis

Let's look into another aspect of running activities: running performances over time. Was there performance difference over time? How can they be measured?

for i in tqdm(range(len(runsTrajectories))):
    runsTrajectories[i].add_speed(overwrite=True)
for i in tqdm(range(len(runsTrajectories))):
    runsTrajectories[i].df['kmh'] = runsTrajectories[i].df['speed'].apply(utils.ms_to_km)

100%|██████████| 135/135 [00:35<00:00,  3.82it/s]
100%|██████████| 135/135 [00:00<00:00, 578.40it/s]

Getting average pace and runs' lengths

runAvg = pd.DataFrame(columns={'date', 'avgPace', 'length'})
for i in tqdm(range(len(runsTrajectories))):
    runAvg.at[i, 'date'] = runsTrajectories[i].get_start_time()
    runAvg.at[i, 'avgPace'] = runsTrajectories[i].df.kmh.mean()
    runAvg.at[i, 'length'] = runsTrajectories[i].get_length()/1000

100%|██████████| 135/135 [00:32<00:00,  4.10it/s]

runAvg.sort_values(by='date', inplace=True)
runAvg.reset_index(drop=True, inplace=True)
for i in tqdm(range(len(runAvg))):
    runAvg.date[i] = runAvg.date[i].strftime("%Y-%m-%d")
runAvg

100%|██████████| 135/135 [00:00<00:00, 2947.85it/s]

	length	avgPace	date
0	5.80499	10.9627	2019-04-09
1	3.00899	12.725	2019-04-11
2	3.9967	12.8169	2019-04-16
3	3.94465	11.2084	2019-04-17
4	3.01506	12.7177	2019-05-01
...	...	...	...
130	6.45656	10.4523	2021-08-30
131	8.03258	12.7818	2021-09-23
132	10.0129	12.1618	2021-09-27
133	6.60791	11.4554	2021-09-29
134	21.2451	12.1443	2021-10-03

135 rows × 3 columns

Exploiting VO2Max data

VO2Max is a measure of maximum oxygen consumption per minute for any muscle contraction. This is a biological parameter which can be partially improved by trainings. Was I able to modify it through my activities? When did it decrease?

data = []
vo2 = pd.DataFrame()
for file in tqdm(os.listdir("data/vo2/")):
    with open("data/vo2/{}".format(file), 'r') as f:
        data = json.loads([line.strip() for line in f][0])
        vo2 = vo2.append(data)
vo2.reset_index(drop=True, inplace=True)
vo2.drop(columns={'userProfilePK', 'deviceId','subSport', 'activityUuid'}, inplace=True)
vo2

100%|██████████| 10/10 [00:00<00:00, 124.80it/s]

	calendarDate	timestampGmt	sport	activityId	vo2MaxValue
0	2019-03-20	2019-03-20T18:53:32.0	RUNNING	3479925555	53.0
1	2019-03-26	2019-03-26T18:39:07.0	RUNNING	3498156058	53.0
2	2019-04-02	2019-04-02T17:44:54.0	RUNNING	3519671950	53.0
3	2019-04-09	2019-04-09T18:14:30.0	RUNNING	3541479161	54.0
4	2019-04-11	2019-04-11T06:43:22.0	RUNNING	3545420258	56.0
...	...	...	...	...	...
133	2021-05-14	2021-05-14T11:57:57.0	RUNNING	6777359705	52.0
134	2021-05-15	2021-05-15T09:54:00.0	RUNNING	6782369267	51.0
135	2021-05-18	2021-05-18T17:57:59.0	RUNNING	6802088682	51.0
136	2021-05-23	2021-05-23T18:53:58.0	RUNNING	6830626048	51.0
137	2021-05-25	2021-05-25T18:49:31.0	RUNNING	6841608538	52.0

138 rows × 5 columns

Efforts comparison

complete = pd.merge(runAvg, vo2, left_on='date', right_on='calendarDate', how="inner")
complete.drop(columns={'calendarDate'}, inplace=True)
complete

	length	avgPace	date	timestampGmt	sport	activityId	vo2MaxValue
0	5.80499	10.9627	2019-04-09	2019-04-09T18:14:30.0	RUNNING	3541479161	54.0
1	3.00899	12.725	2019-04-11	2019-04-11T06:43:22.0	RUNNING	3545420258	56.0
2	3.9967	12.8169	2019-04-16	2019-04-16T17:51:35.0	RUNNING	3561796047	56.0
3	3.94465	11.2084	2019-04-17	2019-04-17T17:34:43.0	RUNNING	3564901233	56.0
4	3.01506	12.7177	2019-05-01	2019-05-01T18:48:20.0	RUNNING	3609227182	56.0
...	...	...	...	...	...	...	...
118	5.01024	13.2755	2021-08-14	2021-08-14T18:45:04.0	RUNNING	7304989718	52.0
119	8.03258	12.7818	2021-09-23	2021-09-23T18:48:04.0	RUNNING	7541750132	53.0
120	10.0129	12.1618	2021-09-27	2021-09-27T18:26:23.0	RUNNING	7564463452	51.0
121	6.60791	11.4554	2021-09-29	2021-09-29T17:38:44.0	RUNNING	7575641530	52.0
122	21.2451	12.1443	2021-10-03	2021-10-03T09:46:02.0	RUNNING	7594999558	51.0

123 rows × 7 columns

utils.plotRunComparison(complete)

There are some patterns that could be noticed:

• activities' length increased over time, but the average pace didn't change too much (except for some isolated activities)
• VO2Max had a low point after the 2020 spring lockdown, as predictable
• interestingly, autumn 2020 and winter 2021 were the period in which I was more trained (according to VO2Max); in fact, in that period activities' length started to raise, but the average pace didn't decreased